COVID-19 vaccine effectiveness against severe COVID-19 requiring oxygen therapy, invasive mechanical ventilation, and death in Japan: A multicenter case-control study (MOTIVATE study).

INTRODUCTION: Since the SARS-CoV-2 Omicron variant became dominant, assessing COVID-19 vaccine effectiveness (VE) against severe disease using hospitalization as an outcome became more challenging due to incidental infections via admission screening and variable admission criteria, resulting in a wide range of estimates. To address this, the World Health Organization (WHO) guidance recommends the use of outcomes that are more specific to severe pneumonia such as oxygen use and mechanical ventilation. METHODS: A case-control study was conducted in 24 hospitals in Japan for the Delta-dominant period (August-November 2021; "Delta") and early Omicron (BA.1/BA.2)-dominant period (January-June 2022; "Omicron"). Detailed chart review/interviews were conducted in January-May 2023. VE was measured using various outcomes including disease requiring oxygen therapy, disease requiring invasive mechanical ventilation (IMV), death, outcome restricting to "true" severe COVID-19 (where oxygen requirement is due to COVID-19 rather than another condition(s)), and progression from oxygen use to IMV or death among COVID-19 patients. RESULTS: The analysis included 2125 individuals with respiratory failure (1608 cases [75.7%]; 99.2% of vaccinees received mRNA vaccines). During Delta, 2 doses provided high protection for up to 6 months (oxygen requirement: 95.2% [95% CI:88.7-98.0%] [restricted to "true" severe COVID-19: 95.5% {89.3-98.1%}]; IMV: 99.6% [97.3-99.9%]; fatal: 98.6% [92.3-99.7%]). During Omicron, 3 doses provided high protection for up to 6 months (oxygen requirement: 85.5% [68.8-93.3%] ["true" severe COVID-19: 88.1% {73.6-94.7%}]; IMV: 97.9% [85.9-99.7%]; fatal: 99.6% [95.2-99.97]). There was a trend towards higher VE for more severe and specific outcomes. CONCLUSION: Multiple outcomes pointed towards high protection of 2 doses during Delta and 3 doses during Omicron. These results demonstrate the importance of using severe and specific outcomes to accurately measure VE against severe COVID-19, as recommended in WHO guidance in settings of intense transmission as seen during Omicron.

Machine learning algorithms for predicting days of high incidence for out-of-hospital cardiac arrest.

Predicting out-of-hospital cardiac arrest (OHCA) events might improve outcomes of OHCA patients. We hypothesized that machine learning algorithms using meteorological information would predict OHCA incidences. We used the Japanese population-based repository database of OHCA and weather information. The Tokyo data (2005-2012) was used as the training cohort and datasets of the top six populated prefectures (2013-2015) as the test. Eight various algorithms were evaluated to predict the high-incidence OHCA days, defined as the daily events exceeding 75% tile of our dataset, using meteorological and chronological values: temperature, humidity, air pressure, months, days, national holidays, the day before the holidays, the day after the holidays, and New Year's holidays. Additionally, we evaluated the contribution of each feature by Shapley Additive exPlanations (SHAP) values. The training cohort included 96,597 OHCA patients. The eXtreme Gradient Boosting (XGBoost) had the highest area under the receiver operating curve (AUROC) of 0.906 (95% confidence interval; 0.868-0.944). In the test cohorts, the XGBoost algorithms also had high AUROC (0.862-0.923). The SHAP values indicated that the "mean temperature on the previous day" impacted the most on the model. Algorithms using machine learning with meteorological and chronological information could predict OHCA events accurately.